ledmands
commited on
Commit
•
b7f0280
1
Parent(s):
5d0eb73
Updated ReadMe
Browse files
README.md
CHANGED
|
@@ -10,7 +10,26 @@ tags:
|
|
| 10 |
|
| 11 |
# Agent using DQN to play ALE/Pacman-v5
|
| 12 |
|
|
|
|
|
|
|
| 13 |
This is an agent that is trained using Stable Baselines3 as part of the capstone project for South Hills School in Spring 2024.
|
| 14 |
The goal of this project is to gain familiarity with reinforcement learning concepts and tools, and to train an agent to score up into the 400-500 point range in Pacman.
|
| 15 |
|
| 16 |
-
##
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
|
| 11 |
# Agent using DQN to play ALE/Pacman-v5
|
| 12 |
|
| 13 |
+
## UPDATE 16 May 2024: Latest DQN model is version 2.6.
|
| 14 |
+
|
| 15 |
This is an agent that is trained using Stable Baselines3 as part of the capstone project for South Hills School in Spring 2024.
|
| 16 |
The goal of this project is to gain familiarity with reinforcement learning concepts and tools, and to train an agent to score up into the 400-500 point range in Pacman.
|
| 17 |
|
| 18 |
+
## Description of Python scripts
|
| 19 |
+
To run a script, first ensure that Python is installed. From the root directory of the repository, run python <script_name> <options>.
|
| 20 |
+
For a list of available options, run python <script_name> --help.
|
| 21 |
+
### watch_agent.py
|
| 22 |
+
This will render the specified agent in real-time.
|
| 23 |
+
Does not save any evaluation information.
|
| 24 |
+
### evaluate_agent.py
|
| 25 |
+
This will evaluate a specified agent and append the results to a specified log file.
|
| 26 |
+
### get_config.py
|
| 27 |
+
This will pull configuration information from the specified agent and save it in JSON format.
|
| 28 |
+
### record_video.py
|
| 29 |
+
This will record a video of a specified agent being evaluated.
|
| 30 |
+
Does not save any evaluation information.
|
| 31 |
+
Currently in major development.
|
| 32 |
+
### plot_evaluations.py
|
| 33 |
+
This will plot the evaluation data that was gathered during the training run of the specified agent using MatPlotLib.
|
| 34 |
+
Charts can be saved to a directory of the user's choosing.
|
| 35 |
+
Currently in major development.
|